The goal of this study is to examine the relationship between state-level Marijuana Expansion Policy (MEP) changes and trends in adolescent marijuana and alcohol use in adolescents. This study will: (1) evaluate available data resources and prepare a dataset containing the necessary variables for analyses, (2) determine adolescent marijuana, alcohol and alcohol/marijuana co-use estimates for 1993-2011, and (3) assess: (a) the effect of MEP(a) and MEP components on trends in report of marijuana, alcohol, and co-use and (b) MEP components that associate with trends in report of marijuana, alcohol, and co-use. The primary marijuana and alcohol outcomes of interest are: 1) lifetime use prevalence and quantity; 2) past-30-day use prevalence and quantity; 3) age of initiation; and 4) past 30-day concurrent alcohol and marijuana use prevalence. Descriptive analyses, factor analyses and multilevel modeling will be employed in this study. Once all variables are constructed, the YRBSS specific marijuana, alcohol, and co-use trends will be examined using univariate statistics and bivariate analyses. As part of variable construction, factor analyses wil be used to construct one or more factors categorizing the various components of decriminalization and medical marijuana policies and tested as alternatives to individual MEPs indicators in the models. Multilevel Modeling (MLM) will be used to control for correlation among clusters and complex patterns of variability within the hierarchy of individual adolescents nested within both states and years. Multilevel modeling assists in the exploration of: (1) relationships at multiple levels of analysis and (2) cross-level interactions. It is theorized that MEPs will afect adolescent marijuana consumption by influencing price, availability, norms, perceptions, and social sanctions. The study hypothesizes that changes in state MEP(s) will be associated with higher report of adolescent marijuana use. A better understanding how varying MEP(s) may affect adolescent marijuana, alcohol, and co-use will help us understand when and how best to prevent and intervene effectively. The results from this study will assist to identify MEP(s) and MEPs policy components that are associated with adolescent marijuana, alcohol and co-use use. This knowledge will lend itself to the development of targeted prevention and interventions. Training under this fellowship would build on Ms. Johnson's previous methodological and analytical skills cultivated in coursework and related research experience to ensure rigorous and timely study completion. The training plan will build develop and hone expertise in: (1) working with secondary data (e.g. assessment of YRBSS datasets, missing data etc.), (2) data management (e.g. constructing necessary variables, merging multiple datasets etc.), (3) quantitative analytical methods (multilevel modeling, factor analyses), (4) theory and (5) research dissemination (e.g. writing abstracts and manuscripts for professional meetings and manuscript publication, networking with researchers etc.).